// ⌖ Interpretability

Grad-CAM

Gradient-weighted heatmaps showing where a CNN looks.

Grad-CAM 热力图 2016 Foundational extended

IN PLAIN WORDS

Weight the last conv layer's channels by the class gradient to get a coarse localization heatmap, revealing whether the model looks at the right place.

Used in: Debugging medical and vision models

TIMELINE · prerequisites → this → what builds on it

1986 Backpropagati…1998 Convolutional…2016 Grad-CAM

01Original paper

Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization

Ramprasaath R. Selvaraju, Michael Cogswell, Abhishek Das, et al. · 2016

arXiv ↗ ✓ Verified · arXiv API

02Where it sits

Learn the left column first; the right column builds on this one.

Learn first

Grad-CAM

Then you can learn

Nothing in the atlas builds on it yet

Left: prerequisites · right: builds on this · dashed: related. Full map →

03Learning resources

Official docs, explainers, courses and reference code. All links checked.

Reference code

jacobgil/pytorch-grad-cam ↗EN

Official docs

captum.ai ↗EN

04Where the course teaches it

U91 Interpretability: from Attribution to Circuits Attribution, probes, the logit lens, superposition, SAEs, circuit tracing, knowledge editing and activation steering. 10 h9 algorithms4 paperslesson ready

For AI / data

Verification: verified, checked 2026-10-21.
JSON: /api/learn/algorithms.json (id: grad-cam)